Author Topic: FLATS !  (Read 8816 times)

Offline zaubermantel

  • Newcomer
  • Posts: 1
    • View Profile
« Reply #45 on: 2016 July 26 11:19:04 »
Interesting -- I have this same problem with my flats. I solved it over the weekend by doing manual flat calibration, but dividing by the flat value plus an offset rather than just the flat value. So:

sub / (flat + 0.6) worked great whereas

sub / flat gave me bright regions at the corners.

I guess this is equivalent to doing a linear fit of the flat to the sub before calibrating, right? I will try Geert's method when I get a chance but I suspect it will work as well or better.

Offline gvanhau

  • PixInsight Old Hand
  • ****
  • Posts: 345
    • View Profile
« Reply #46 on: 2016 July 26 19:09:25 »

Here is the link to the proyect (zipped file)

The whole proyect uses drive F to read an store files.
On screen, from left to right:
I integrated dark files, Then integrated FlatDark files
Then I calibrated flats only using master FlatDark (bias is already included in master FlatDark).
Then I Integrated calibrated flats, giving a normal masterflat.

The next step is my proposal:
Y use splitCFA to split a single light, The master Dark and The master Flat
Then with pixelMath I generate a semi calibrated light for each of the CFA0...CFA3
(note the cfa images ar written in separate folders into the f drive)
the resulting images are then used to linear fit each of the CFA images of the Master Flat.
Finally the resulting CFA_Flats  are recombined into one image using merge CFA.
This resulting master flat  is then used to calibrate the lights.

Note: calibration of lights is done only using MAsterDark and the master flat from previous steps.
No bias is necesary because it is already included into the masterDark.

note: Batch debayer is a placeholder for the batch debayer script.

Hope this litle explanation hepls.


Geert Vanhauwaert